Fusion SLAM Map Updating With LiDAR-Camera Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robots using single sensors for SLAM (Simultaneous Localization and Mapping) face challenges in accurately updating maps and maintaining alignment between different sensor data, leading to reduced map accuracy and efficiency in navigation.
Innovation Solution
A method employing two types of sensors, such as LiDAR and camera sensors, where one sensor type is used for localization and the other for updating the map, ensuring accurate position estimation and map updating by maintaining alignment between different sensor data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a robot uses a single sensor for SLAM, then the device complexity is reduced, but the map accuracy and localization precision deteriorate
Solution Approach 1:
The patent combines multiple sensor types (LiDAR and camera) into a unified SLAM system. The LiDAR sensor generates a LiDAR map while the camera sensor generates a visual map, and both maps are fused together to create a more accurate and reliable representation of the environment. This merging of sensor data resolves the contradiction by achieving high map accuracy through multi-sensor integration while managing device complexity through systematic data fusion.
Solution Approach 2:
The patent creates a composite mapping system where LiDAR data and visual data are combined to form a hybrid map representation. The LiDAR map provides structural and geometric information while the visual map provides texture and color information, creating a composite map that leverages the strengths of both sensor types to achieve superior accuracy compared to single-sensor systems.
2Measurement precision
If a robot uses multiple sensors for SLAM, then the map accuracy is enhanced, but the device complexity increases
Solution Approach 1:
The patent segments the mapping process into distinct components: LiDAR-based mapping and visual-based mapping. Each sensor type processes its own data independently to create separate maps (LiDAR map and visual map), which are then integrated. This segmentation approach manages device complexity by organizing the complex multi-sensor system into manageable, independent processing streams that can be systematically combined.
Solution Approach 2:
The patent implements a universal mapping framework that can process and integrate data from different sensor types (LiDAR and camera) using a common SLAM algorithm structure. The system maintains both a LiDAR map and a visual map using the same underlying technology platform, allowing the system to handle multiple sensor inputs through a unified processing architecture, thereby managing complexity while achieving high accuracy.
3Stability of the object's composition
If a robot maintains alignment between different sensor data, then the map consistency is improved, but the processing time increases
Solution Approach 1:
The patent implements a self-aligning mapping system where the LiDAR map and visual map automatically maintain alignment through their shared coordinate system and synchronized processing. The system inherently preserves consistency between different map representations by using the same spatial reference framework, eliminating the need for separate alignment operations and reducing processing time while maintaining map consistency.
Data Source
AI summary
Disclosed herein are a method of updating a map in fusion SLAM and a robot implementing the same, the robot, which updates a map in fusion SLAM using two types of sensors, configured to update a first map with first type information acquired by a first sensor and to estimate a current position of the robot using second type information acquired by a second sensor.